Automatic Trash Classification using Convolutional Neural Network Machine Learning

Patipol Tiyajamorn, Pollakrit Lorprasertkul, Rawin Assabumrungrat, Warin Poomarin, Ratchatin Chancharoen

Abstract: In our current society, there is an increasing amount of trash generated every day. In Thailand, this has become a severe problem, as there are more than 27 million tons of trash that cannot be processed in time generated each year due to the current way of processing garbage. This paper proposes a potential solution by creating AlphaTrash, a machine that can be fitted to conventional curb-side trashcan and used to sort out deposit trash automatically. By using a retrained convolutional neural network (Inception-v1), the machine can classify trash with an accuracy of 94%, while using 4.2 seconds per classification.